Convolutional Neural Networks¶
Project: Write an Algorithm for a Dog Identification App¶
In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
Why We're Here¶
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).
In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
The Road Ahead¶
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
- Step 0: Import Datasets
- Step 1: Detect Humans
- Step 2: Detect Dogs
- Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
- Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
- Step 5: Write your Algorithm
- Step 6: Test Your Algorithm
Step 0: Import Datasets¶
Make sure that you've downloaded the required human and dog datasets:
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location
/dogImages.Download the human dataset. Unzip the folder and place it in the home directory, at location
/lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("lfw/*/*"))
dog_files = np.array(glob("dogImages/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
Step 1: Detect Humans¶
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
Write a Human Face Detector¶
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
(IMPLEMENTATION) Assess the Human Face Detector¶
Question 1: Use the code cell below to test the performance of the face_detector function.
- What percentage of the first 100 images in
human_fileshave a detected human face? - What percentage of the first 100 images in
dog_fileshave a detected human face?
Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: (You can print out your results and/or write your percentages in this cell)
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
detected_in_human = sum(map(face_detector, human_files_short))
detected_in_dog = sum(map(face_detector, dog_files_short))
f'Faces detected in {detected_in_human}% of the short human dataset and {detected_in_dog}% of the dog dataset'
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
Step 2: Detect Dogs¶
In this section, we use a pre-trained model to detect dogs in images.
Obtain Pre-trained VGG-16 Model¶
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
use_cuda
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
(IMPLEMENTATION) Making Predictions with a Pre-trained Model¶
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
# Set PIL to be tolerant of image files that are truncated.
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def load_image(img_path, img_size):
tfms = transforms.Compose([
transforms.Resize(img_size),
transforms.CenterCrop(img_size),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
img_tensor = tfms(Image.open(img_path))
if use_cuda:
img_tensor = img_tensor.cuda()
return img_tensor
def predict_with_model(img_path, model, img_size = 224):
'''
Use a given model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
model: the model to use for prediction
Returns:
Index corresponding to the model's prediction
'''
img_tensor = load_image(img_path, img_size)
model.eval()
classes = model(img_tensor[None])
max_idx = torch.argmax(classes)
return max_idx.item() # predicted class index
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
return predict_with_model(img_path, VGG16)
Lets plot some of the images with corresponding labels to see if it works
import pickle
import urllib
dict_url = 'https://gist.githubusercontent.com/yrevar/6135f1bd8dcf2e0cc683/raw/d133d61a09d7e5a3b36b8c111a8dd5c4b5d560ee/imagenet1000_clsid_to_human.pkl'
imagenet_class_dict = pickle.load(urllib.request.urlopen(dict_url) )
_, axs = plt.subplots(4,4,figsize=(16,16))
for ax, img in zip(axs.flatten(), dog_files_short):
ax.imshow(Image.open(img))
pred_idx = VGG16_predict(img)
ax.set_title(imagenet_class_dict[pred_idx])
ax.axis('off')
(IMPLEMENTATION) Write a Dog Detector¶
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def is_dog_index(class_idx):
return class_idx >= 151 and class_idx <= 268
def dog_detector(img_path):
return is_dog_index(VGG16_predict(img_path))
(IMPLEMENTATION) Assess the Dog Detector¶
Question 2: Use the code cell below to test the performance of your dog_detector function.
- What percentage of the images in
human_files_shorthave a detected dog? - What percentage of the images in
dog_files_shorthave a detected dog?
Answer:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
detected_in_human = sum(map(dog_detector, human_files_short))
detected_in_dog = sum(map(dog_detector, dog_files_short))
f'Dogs detected in {detected_in_human}% of the short human dataset and {detected_in_dog}% of the dog dataset'
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
resnet = models.resnet50(pretrained = True)
if use_cuda:
resnet = resnet.cuda()
resnet_predictor = lambda img: is_dog_index(predict_with_model(img, resnet))
detected_in_human = sum(map(resnet_predictor, human_files_short))
detected_in_dog = sum(map(resnet_predictor, dog_files_short))
f'Dogs detected with ResNet50 in {detected_in_human}% of the short human dataset and {detected_in_dog}% of the dog dataset'
Step 3: Create a CNN to Classify Dog Breeds (from Scratch)¶
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
 |
 |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
 |
 |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
 |
 |
 |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import torch
import torchvision
from torchvision import datasets
import numpy as np
batch_size = 32
img_size = 128
train_tfms = transforms.Compose([
transforms.Resize(img_size),
transforms.ColorJitter(),
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.RandomRotation(10),
transforms.CenterCrop(img_size),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
test_tfms = transforms.Compose([
transforms.Resize(img_size),
transforms.CenterCrop(img_size),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
train_data = torchvision.datasets.ImageFolder('dogImages/train', transform = train_tfms)
val_data = torchvision.datasets.ImageFolder('dogImages/valid', transform = test_tfms)
test_data = torchvision.datasets.ImageFolder('dogImages/test', transform = test_tfms)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=8)
val_loader = torch.utils.data.DataLoader(val_data, batch_size=batch_size, shuffle=True, num_workers=8)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, shuffle=True, num_workers=8)
loaders_scratch = {'train': train_loader, 'valid' : val_loader, 'test' : test_loader}
Question 3: Describe your chosen procedure for preprocessing the data.
- How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
- Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?
Answer: I first resize the images to have the smaller dimension equal to 128 and then center crop along the longer dimension to make it square. This way there is no stretching, though there is a risk of losing part of an image that could have been useful. I picked the size 128 because I wanted a small size to make training and experimentation fast. I started with 224 as in imagenet, but then reduced it to make it faster and it still gave me ok results.
I'm augmenting the training set with horizontal flip, color jitter and a small random rotation (max 10 degrees) - this reduced my overfitting a bit.
(IMPLEMENTATION) Model Architecture¶
Create a CNN to classify dog breed. Use the template in the code cell below.
no_classes = len(train_data.classes)
no_classes
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
## Define layers of a CNN
channels = [3, 16, 32, 64, 128, 256]
self.convolutions = nn.Sequential()
for idx in range(1, len(channels)):
self.convolutions.add_module(f'conv_{idx}',
nn.Conv2d(channels[idx-1], channels[idx], kernel_size=3,padding=1))
self.convolutions.add_module(f'bn_{idx}', nn.BatchNorm2d(channels[idx]))
self.convolutions.add_module(f'relu_{idx}', nn.ReLU())
self.convolutions.add_module(f'maxpool_{idx}', nn.MaxPool2d(kernel_size=2, stride=2))
self.classifier = nn.Sequential(
nn.Linear(channels[-1], 256),
nn.ReLU(True),
nn.Dropout(),
nn.Linear(256, 256),
nn.ReLU(True),
nn.Dropout(),
nn.Linear(256, no_classes),
)
def forward(self, x):
## Define forward behavior
x = self.convolutions(x)
x = F.adaptive_max_pool2d(x, 1).view(x.shape[0],-1)
x = self.classifier(x)
return x
#-#-# You do NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
model_scratch
Verify that I can pass a batch of images without an error and get the correct output shape:
b_x, b_y = next(train_loader.__iter__())
model_scratch(b_x.cuda()).shape
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer:
I wanted to start with something as simple and small as possible to make training and iterating on the design faster. Each layer is a block of a 2d convolution and batchnorm followed by the relu activation and maxpooling. So it's very much like the VGG architecture, the number of layers and features is decided by the Net.channels list so I can modify the architecture without changing any other code.
For the classifier I first do adaptive max pool, to get one value out of each channel, since I only specify the desired output size this works regardless of the initial image size and the number of layers. After that I have three linear layers and I tried to make them as small as possible for fast training.
I first tried only 5 convolutional layers and 128 neurons in each linear layer - but it was giving me accuracy barely over 10%. I added one more layer and doubled the size of linear layers and got to 33%.
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()
### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters(),lr=0.001)
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
# the following import is required for training to be robust to truncated images
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders['train']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
output = model(data)
optimizer.zero_grad()
lss = criterion(output, target)
lss.backward()
optimizer.step()
train_loss += data.shape[0] * lss
## find the loss and update the model parameters accordingly
## record the average training loss, using something like
## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
train_loss /= len(loaders['train'].dataset)
######################
# validate the model #
######################
model.eval()
with torch.no_grad():
for batch_idx, (data, target) in enumerate(loaders['valid']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
## update the average validation loss
output = model(data)
lss = criterion(output, target)
valid_loss += data.shape[0] * lss
valid_loss /= len(loaders['valid'].dataset)
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch,
train_loss,
valid_loss
))
if valid_loss < valid_loss_min:
valid_loss_min = valid_loss
print('saving at valid loss', valid_loss.item())
torch.save(model.state_dict(), save_path)
# return trained model
return model
%%time
# train the model
model_scratch = train(50, loaders_scratch, model_scratch, optimizer_scratch,
criterion_scratch, use_cuda, 'model_scratch.pt')
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}\n'.format(test_loss))
print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
100. * correct / total, correct, total))
# call test function
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)¶
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset¶
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
## using the same loaders as before
loaders_transfer = {'train': train_loader, 'valid' : val_loader, 'test' : test_loader}
(IMPLEMENTATION) Model Architecture¶
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
import torchvision.models as models
import torch.nn as nn
model_transfer = models.vgg11_bn(pretrained=True)
model_transfer.classifier = nn.Sequential(
nn.Linear(8192, 256),
nn.ReLU(True),
nn.Dropout(),
nn.Linear(256, 256),
nn.ReLU(True),
nn.Dropout(),
nn.Linear(256, no_classes),
)
for param in model_transfer.features.parameters():
param.requires_grad = False
if use_cuda:
model_transfer = model_transfer.cuda()
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer:
I want to have something similar to my first network since I know it worked fine and want to compare the results between them. I'm using the vgg_11 the smallest out of vggs available (though still much larger than my baby network).
I'm using the same classifier as I did before, but instead the average max pool I funnel all the outputs into the first linear layer since that's what the vgg does.
(IMPLEMENTATION) Specify Loss Function and Optimizer¶
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.Adam(model_transfer.classifier.parameters(),lr=1e-3)
(IMPLEMENTATION) Train and Validate the Model¶
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# train the model by modyfing the classifier only
model_transfer = train(10, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer_cl.pt')
# load the model that got the best validation accuracy
model_transfer.load_state_dict(torch.load('model_transfer_cl.pt'))
Unfreeze all the layers and train some more.
for param in model_transfer.features.parameters():
param.requires_grad = True
optimizer_transfer_all = optim.Adam(model_transfer.parameters(),lr=1e-4)
model_transfer = train(20, loaders_transfer, model_transfer, optimizer_transfer_all, criterion_transfer, use_cuda, 'model_transfer.pt')
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
(IMPLEMENTATION) Test the Model¶
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
(IMPLEMENTATION) Predict Dog Breed with the Model¶
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in loaders_transfer['train'].dataset.classes]
def predict_breed_transfer(img_path):
idx = predict_with_model(img_path, model_transfer, img_size = 128)
return class_names[idx]
Plotting some to check if it looks reasonable
_, axs = plt.subplots(4,4,figsize=(16,16))
for ax, img in zip(axs.flatten(), dog_files_short):
ax.imshow(Image.open(img))
predicted = predict_breed_transfer(img)
ax.set_title(predicted)
ax.axis('off')
Step 5: Write your Algorithm¶
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
- if a dog is detected in the image, return the predicted breed.
- if a human is detected in the image, return the resembling dog breed.
- if neither is detected in the image, provide output that indicates an error.
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!
(IMPLEMENTATION) Write your Algorithm¶
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.
def run_app(img_path):
is_face = face_detector(img_path)
is_dog = dog_detector(img_path)
if is_face or is_dog:
breed = predict_breed_transfer(img_path)
plt.imshow(Image.open(img_path))
if is_face:
plt.title(f"It's a human! though it looks a bit like a {breed}")
elif is_dog:
plt.title(f"It's a dog! {breed} to be specific")
else:
# display the imagenet class if it's neither face nor dog
pred_idx = VGG16_predict(img_path)
thing = imagenet_class_dict[pred_idx]
plt.title(f"I don't know what this is, maybe a '{thing}'")
plt.axis('off')
plt.show()
Step 6: Test Your Algorithm¶
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
(IMPLEMENTATION) Test Your Algorithm on Sample Images!¶
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer:
- 74% accuracy is lower than I expected, but I was designing it for fast and simple. We could get it above 90 by using a better pretrained model (like resnet152 for example), higher size images and training it more.
- We use two separate models to detect if something is a dog and then what dog it is, it's possible that the general image net model wouldn't recognize a breed that our more specialized classifier handles correctly. We could fix that by adding and unknown class and some non dog images to the training.
- When comparing human faces to dogs we pass the whole image to the dog classifier, which might then use other parts of the image to determine the breed, not necessarily using the face at all. What we could do instead is find the bounding box of the face first and then zoom it in and use only that as the input for the second model.
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.
from pathlib import Path
for file in Path('my_test_images').iterdir():
run_app(str(file))
